Update src/streamlit_app.py

src/streamlit_app.py

@@ -1,40 +1,140 @@
-import altair as alt
-import numpy as np
-import pandas as pd
+# wound_agent_streamlit.py
+
 import streamlit as st
+import numpy as np
+import cv2
+import torch
+import tempfile
+from PIL import Image
+from tensorflow.keras.models import load_model
+from transformers import pipeline
+
+st.set_page_config(page_title="SmartHeal Wound Agent", layout="wide")
+
+# ------------------------------
+# Load All Models Once
+# ------------------------------
+@st.cache_resource
+def load_all_models():
+    # YOLOv5 detection
+    detection_model = torch.hub.load("ultralytics/yolov5", "custom", path="best.pt", force_reload=False)
+
+    # Segmentation
+    segmentation_model = load_model("segmentation model.h5", compile=False)
+
+    # Classification
+    classification_pipe = pipeline("image-classification", model="Hemg/Wound-classification")
+
+    # Med-Gemma
+    medgemma_pipe = pipeline(
+        "image-text-to-text",
+        model="google/medgemma-4b-it",
+        torch_dtype=torch.bfloat16,
+        device="cuda"
+    )
+
+    return detection_model, segmentation_model, classification_pipe, medgemma_pipe
+
+yolo_model, seg_model, classify_pipe, medgemma = load_all_models()
+
+# ------------------------------
+# Area Estimation
+# ------------------------------
+def estimate_area(mask, px_per_cm=20):
+    pixel_area = np.sum(mask > 0)
+    area_cm2 = pixel_area / (px_per_cm ** 2)
+    return round(area_cm2, 2)
+
+# ------------------------------
+# Main UI
+# ------------------------------
+st.title("🩹 SmartHeal: Real-Time Wound Care Agent")
+
+uploaded_file = st.file_uploader("📤 Upload a wound image", type=["jpg", "jpeg", "png"])
+with st.form("patient_form"):
+    age = st.number_input("Patient Age", min_value=1, max_value=120)
+    diabetic = st.radio("Is the patient diabetic?", ["Yes", "No"])
+    infection = st.radio("Signs of infection present?", ["Yes", "No"])
+    submitted = st.form_submit_button("🔍 Analyze")
+
+if uploaded_file and submitted:
+    image = Image.open(uploaded_file).convert("RGB")
+    st.image(image, caption="Uploaded Image", use_column_width=True)
+
+    # Convert to OpenCV format
+    image_cv = np.array(image)
+
+    # ------------------ DETECTION ------------------
+    st.subheader("🧠 Detection")
+    results = yolo_model(image_cv)
+    boxes = results.xyxy[0].cpu().numpy()
+
+    if len(boxes) == 0:
+        st.error("No wound detected.")
+        st.stop()
+
+    x1, y1, x2, y2 = map(int, boxes[0][:4])
+    detected_region = image_cv[y1:y2, x1:x2]
+
+    # ------------------ SEGMENTATION ------------------
+    st.subheader("🧠 Segmentation")
+    resized = cv2.resize(detected_region, (256, 256)) / 255.0
+    input_tensor = np.expand_dims(resized, axis=0)
+    pred_mask = seg_model.predict(input_tensor)[0]
+    binary_mask = (pred_mask[:, :, 0] > 0.5).astype(np.uint8)
+    mask_resized = cv2.resize(binary_mask, (x2 - x1, y2 - y1))
+    full_mask = np.zeros(image_cv.shape[:2], dtype=np.uint8)
+    full_mask[y1:y2, x1:x2] = mask_resized
+
+    area_cm2 = estimate_area(full_mask)
+    st.markdown(f"📏 **Estimated Wound Area:** `{area_cm2} cm²`")
+
+    overlay = image_cv.copy()
+    overlay[full_mask > 0] = [255, 0, 0]
+    st.image(overlay, caption="Wound Segmentation Overlay", use_column_width=True)
+
+    # ------------------ CLASSIFICATION ------------------
+    st.subheader("🧠 Wound Classification")
+    tmp = tempfile.NamedTemporaryFile(delete=False, suffix=".png")
+    Image.fromarray(detected_region).save(tmp.name)
+    wound_type = classify_pipe(tmp.name)[0]["label"]
+    st.success(f"✅ Wound Type Classified: **{wound_type}**")
+
+    # ------------------ MED-GEMMA ------------------
+    st.subheader("🧠 Med-Gemma Diagnosis + Treatment Plan")
+
+    messages = [
+        {
+            "role": "system",
+            "content": [{"type": "text", "text": "You are a wound care expert."}]
+        },
+        {
+            "role": "user",
+            "content": [
+                {"type": "text", "text": f"""Patient Info:
+- Age: {age}
+- Diabetic: {diabetic}
+- Wound Type: {wound_type}
+- Area: {area_cm2} cm²
+- Signs of infection: {infection}
+
+Please provide:
+1. Wound assessment
+2. Recommended treatment
+3. Cleaning & dressing method
+4. Red flags to monitor
+5. Follow-up schedule"""},
+                {"type": "image", "image": image}
+            ]
+        }
+    ]
+
+    with st.spinner("Generating treatment plan..."):
+        output = medgemma(text=messages, max_new_tokens=300)
+        response = output[0]["generated_text"][-1]["content"]
+
+    st.markdown("### 📝 Recommendation")
+    st.info(response)
+
+    st.download_button("📄 Download Report", response, file_name="treatment_plan.txt")
 
-"""
-# Welcome to Streamlit!
-
-Edit `/streamlit_app.py` to customize this app to your heart's desire :heart:.
-If you have any questions, checkout our [documentation](https://docs.streamlit.io) and [community
-forums](https://discuss.streamlit.io).
-
-In the meantime, below is an example of what you can do with just a few lines of code:
-"""
-
-num_points = st.slider("Number of points in spiral", 1, 10000, 1100)
-num_turns = st.slider("Number of turns in spiral", 1, 300, 31)
-
-indices = np.linspace(0, 1, num_points)
-theta = 2 * np.pi * num_turns * indices
-radius = indices
-
-x = radius * np.cos(theta)
-y = radius * np.sin(theta)
-
-df = pd.DataFrame({
-    "x": x,
-    "y": y,
-    "idx": indices,
-    "rand": np.random.randn(num_points),
-})
-
-st.altair_chart(alt.Chart(df, height=700, width=700)
-    .mark_point(filled=True)
-    .encode(
-        x=alt.X("x", axis=None),
-        y=alt.Y("y", axis=None),
-        color=alt.Color("idx", legend=None, scale=alt.Scale()),
-        size=alt.Size("rand", legend=None, scale=alt.Scale(range=[1, 150])),
-    ))